JP5384452B2 - Distribution server and recommendation method - Google Patents

Description

  The present invention relates to a distribution server and method for recommending an application to a user.

  Recent communication terminals not only perform communication, but also can download and execute various applications from a server. Specifically, the communication terminal is a mobile phone, an information terminal, a smartphone, a notebook personal computer, a laptop computer, a desktop computer, or the like, but is not limited thereto. Since there are an extremely large number of applications that can be downloaded to the communication terminal, the application provider (provider) needs to recommend the application to be provided to the user. About recommending an application to the user of a communication terminal is described in Patent Document 1, for example.

JP 2010-157207 A

  Applications existing on the market range from applications that can be used relatively easily such as applications for e-mail and applications for browsers to complex applications that have many advanced functions. In this case, it is conceivable to recommend to the user A an application owned by the user B similar to the user A among the applications not owned by the user A. However, when an application to be recommended is determined by such a simple method, there is a concern that a complicated and advanced application is recommended to a beginner who has little experience. Or, conversely, there is a concern that an application having only a simple function is recommended to an experienced veteran user. As a result, there is a concern that useful application information cannot be efficiently provided to the user.

  An object of the present invention is to enable a server that distributes an application to recommend the application to the user in consideration of the proficiency level of the user.

A distribution server according to an aspect of the present invention includes:
A receiving unit that receives usage history information indicating the usage history of the application from a plurality of users;
A proficiency level calculating unit for calculating proficiency levels of each of a plurality of users having a specific application from the usage history information;
Classifying the plurality of users having the specific application into two or more groups according to the proficiency level of each of the users, and determining an association that associates the specific application with any group; and
A filtering unit that determines an application recommended for the specific user by determining an application corresponding to a group to which the specific user belongs by the correspondence relationship;
A distribution server for recommending an application to a user, comprising: a transmission unit that transmits a recommendation signal for recommending an application determined to be recommended by the filtering unit to the specific user.

  According to an aspect of the present invention, a server that distributes an application can recommend an application to a user in consideration of the user's proficiency level.

The figure which shows the communication system in an Example. The functional block diagram of a communication terminal. The functional block diagram of a delivery server. The figure which shows typically the distribution function showing the relationship between proficiency and the number of users. The figure which shows a mode that the information of the user (alpha) and the user (beta) was projected on the feature space. The figure for demonstrating the operation | movement outline | summary in an Example. The sequence diagram which shows the operation example in an Example. The figure which shows another distribution function typically. The figure which shows another distribution function typically. The sequence diagram which shows the operation example in a modification. The figure for demonstrating the operation | movement in a modification.

  Examples of the present invention will be described from the following viewpoints.

1. System 2. 2. Communication terminal Distribution server 4. 4. Outline of operation Example of operation Modified example

<1. System>
FIG. 1 shows a communication system 1 used in the embodiment. The communication system 1 includes a communication terminal 10, a mobile communication network 20, a distribution server 30, and an application management server 40. The communication terminal 10 communicates with the distribution server 30 via the mobile communication network 20. The distribution server 30 communicates with the application management server 40 as necessary. The application management server 40 has various applications that can be provided by the provider, and provides applications in response to requests from the distribution server 30. In the illustrated example, the application management server 40 is shown as a network element different from the distribution server 30, but this is not essential, and the application management server 40 exists in the distribution server 30. May be.

<2. Communication terminal>
FIG. 2 shows a functional block diagram of the communication terminal 10. FIG. 2 shows various functional elements included in the communication terminal 10 that are particularly relevant to the present embodiment. The communication terminal may be any appropriate communication terminal that can download and execute an application from the distribution server 30. The communication terminal may be a mobile terminal or a fixed terminal, and specifically includes a mobile phone, an information terminal, a smartphone, a notebook personal computer, a laptop computer, a desktop computer, and the like, but is not limited thereto. The illustrated communication terminal 10 includes at least a transmission / reception unit 202, a terminal operation management unit 204, a resource usage management unit 206, an application management unit 208, and a display unit 210.

  The transmission / reception unit 202 has a function as a reception unit and a function as a transmission unit of the communication terminal. The transmission / reception unit 202 receives an application and other signals from the distribution server via the mobile communication network. Other signals include, for example, a recommendation signal that prompts the user to download some application. The transmission / reception unit 202 transmits a signal to the distribution server via the mobile communication network. The transmitted signal includes, for example, usage history information.

  The usage history information is acquired by a terminal operation management unit 204, a resource usage management unit 206, an application management unit 208, and the like, which will be described later. The usage history information may be reported to the distribution server at a certain frequency, or may be reported in response to a request from the user, the distribution server, or the like. The frequency when reporting the usage history information at a certain frequency may be any appropriate frequency such as time, day, week, month, and the like. From the viewpoint of storing the latest information in the distribution server and reducing the amount of usage history information stored in the communication terminal, it is desirable to report at a high frequency. From the viewpoint of effective use of network resources, it is desirable to report as necessary.

  The terminal operation management unit 204, the resource usage management unit 206, and the application management unit 208 manage and acquire information to be included in the usage history information. The provision of the terminal operation management unit 204, the resource usage management unit 206, and the application management unit 208 in the communication terminal 10 is merely an example, and there is another management unit that acquires and manages information that can be used history information. It may be. The usage history information may include other information such as a place where the application is used in addition to the application name, date and time, usage time, and the like.

  The terminal operation management unit 204 acquires and manages information on the operation amount of the communication terminal. The amount of operation information includes, for example, how long the user has used the communication terminal (time), how often the user interface has been operated (number of times or the number of times per unit time), and the touch panel of the display unit 210. It is expressed by how often it was operated (number of times or number of times per unit time), how often the button of the communication terminal was pressed (number of times or number of times per unit time), etc. Instead, it may be expressed in any appropriate amount. The operation amount information may indicate an amount per unit time. The unit time may be any suitable period such as one day, one week, or one month.

  The resource usage management unit 206 acquires and manages information on the resource usage used in the communication terminal. The resource usage is expressed, for example, by the CPU usage (time) of the communication terminal, the usage of the storage device (bit, information), the communication data (bit, bit / second), the usage time of the GPS receiver, etc. However, the present invention is not limited to these and may be expressed in any appropriate amount. Note that the storage device includes not only a built-in memory built in the communication terminal but also an external memory detachably attached to the communication terminal.

  The application management unit 208 acquires and manages information on applications used in the communication terminal. Application information is based on the date and time when the application was downloaded or saved on the communication terminal, the number of applications downloaded or saved, the number of applications installed on the communication terminal, the date and time when the application was started, the information on the deleted application, etc. Although it may be expressed, it is not limited to these and may be expressed in any appropriate amount.

  The display unit 210 functions as a visual user interface. The display unit 210 is specifically a liquid crystal display (LCD), a plasma display, an organic EL panel, or the like, but is not limited thereto. The display unit 210 in the present embodiment realizes a touch screen having a touch-sensitive transparent panel, and can control the display of the screen by the movement of the user's finger or stylus. The display unit 210 can display characters, graphics, symbols, still images, moving images, and the like. For example, an icon representing an application is displayed on the screen.

  Note that the communication terminal 10 may be provided with not only a visual user interface such as the display unit 210 but also other user interfaces. For example, the communication terminal may include a keypad, a control panel, a keyboard, a touch pad, a mouse, a trackball, a microphone, a speaker, and the like.

<3. Distribution server>
FIG. 3 shows a functional block diagram of the distribution server 30. FIG. 3 shows various functional elements provided in the distribution server 30 that are particularly relevant to the present embodiment. The distribution server 30 searches for an application stored in the application management server 40, recommends the application to the user of the communication terminal 10 via the mobile communication network 20, and provides the application according to an instruction from the user. The illustrated distribution server 30 includes at least a transmission / reception unit 302, a usage history information storage unit 304, a proficiency level calculation unit 306, an application determination unit 308, a recommendation candidate determination unit (recommendation candidate determination unit) 310, and a filtering unit 312.

  The transmission / reception unit 302 has a function as a reception unit and a function as a transmission unit of the distribution server 30. The transmission / reception unit 302 receives a signal from the communication terminal 10 via the mobile communication network 20. The received signal includes, for example, the above usage history information. The usage history information is received from each of a plurality of communication terminals connected to the distribution server 30. For the application decided to be recommended to the user, the transmission / reception unit 302 creates a recommendation signal that recommends downloading of the application, and transmits the recommendation signal to the user.

  In addition to storing parameters, calculation results, and other information used by the distribution server 30, the usage history information storage unit 304 stores usage history information received from each user, in particular, in this embodiment.

  The proficiency level calculation unit 306 calculates the proficiency level of each user based on usage history information from various users. In general, the proficiency level is an amount indicating how familiar the user is with the operation of the communication terminal (i.e., proficiency level), and is an amount that is uniquely calculated for one user. However, the level of proficiency is not unchanged and has a property that varies depending on how the user uses the communication terminal. The proficiency level in this description does not directly indicate how familiar the user is with a specific application, but indicates how familiar the user is with an operation of a communication terminal having one or more applications. However, it is expected that users with a high level of proficiency are accustomed to operating specific applications. As described above, the usage history information includes at least information on the operation amount of the communication terminal, information on the resource usage, and information on the application to be used (information other than this may be included).

  As a first example, even if the proficiency level is determined such that the greater the operation amount by a certain user α, the higher the proficiency level of the user α, and the lower the operation amount, the lower the proficiency level of the user α. Good. More specifically, there are many times when the user uses the communication terminal, the number of times that the user interface has been operated (or the number of times per unit time), or the number of times the button of the communication terminal is pressed (or the number of times per unit time). The proficiency level may be determined so that the proficiency level of the user is higher and the proficiency level is lower as the user is less.

  As a second example, the proficiency level is such that the greater the resource usage at a communication terminal of a user α, the higher the proficiency level of the user α, and the lower the resource usage rate, the lower the proficiency level of the user α. May be determined. More specifically, the more the CPU usage (time) of the communication terminal, the storage device usage (bits, information), the communication data volume (bits, bits / second), or the GPS receiver usage time, The proficiency level of the user may be determined such that the proficiency level of the user is high and the proficiency level is lower as the number is lower.

  As a third example, the proficiency level is such that the greater the usage of an application in a communication terminal of a certain user α, the higher the proficiency level of the user α, and the lower the usage rate, the lower the proficiency level of the user α. May be determined. More specifically, the greater the number of applications downloaded or saved on the communication terminal, the number of applications installed on the communication terminal, the number of times the application was started, etc., the higher the proficiency of the user, the less proficiency the less The proficiency level may be determined to be low.

  The proficiency level is determined based on one or more of various amounts shown in the first to third examples. For example, the proficiency level may be determined based only on the number of applications held in the communication terminal. Alternatively, the proficiency level may be determined by a combination (for example, linear combination) of a plurality of the above-described various amounts. When a plurality of the above various amounts are synthesized, it is necessary to normalize or scale each amount (component). As an example, each of a plurality of amounts (components) that contribute to proficiency is normalized such that the average is 0 and the variance is 1.

  The proficiency level calculation unit 306 obtains a distribution function (y = f (x)) representing the relationship between the proficiency level (x) and the number of users having the proficiency level (y) for a certain proficiency value range. . The distribution function is obtained for each application. In other words, users who have a specific application are selected (a more appropriate method is to select users who have a track record of using a specific application), and how the selected user is in terms of proficiency The distribution function can be obtained by examining whether the distribution is in the range.

  FIG. 4A shows a distribution function (y = f (x) regarding a user who has a certain application A (or who has a track record of using a certain application A) and a relationship between the proficiency level x and the number y of users who have the proficiency level. )) Is shown schematically. This shows how various users having a certain application A are distributed from the viewpoint of proficiency. The proficiency level x can be a continuous value or a discrete value depending on the definition of the proficiency level. In the case of the illustrated example, many users are distributed in the vicinity of a low proficiency level, and it can be seen that this application A is widely accepted by users with low proficiency levels.

  The application determination unit 308 in FIG. 3 classifies a plurality of users into at least two groups according to the proficiency level of each user for each application. Typically, a plurality of users are divided into an elementary group, an intermediate group, and an advanced group, but the number of groups is arbitrary. Grouping may be performed in any suitable manner known in the art. As an example, grouping may be performed by cluster analysis by the Ward method (Ward method). In the case of cluster analysis by the Ward method, grouping is performed based on the degree of similarity (distance in the N-dimensional space) between users in the N-dimensional space (feature space) spanned by components of proficiency (number of applications held, etc.). . N is 1 or more.

  As an example, it is assumed that the proficiency level is expressed by one quantity or component of “the number of applications held in the communication terminal”. In this case, the proficiency level can be expressed in one dimension, and the users can be grouped based on the interval or distance between the users on the proficiency level line.

  As another example, when the proficiency is expressed by two quantities or components of “operation time of the communication terminal” and “number of applications stored in the communication terminal”, the distance in the two-dimensional feature space regarding the proficiency Based on the above, users can be grouped. For example, it is assumed that the operation time and the number of applications of the user α are 1 hour and 10 respectively, and the operation time and the number of applications of the user β are 7 hours and 5 respectively.

FIG. 4B shows a state in which the information on the users α and β is projected on a two-dimensional feature space. In this case, the similarity between the user α and the user β is expressed by the distance between the users in the two-dimensional feature space. The distance may be expressed as an Euclidean distance as an example, but the distance may be defined by other methods. In the case of the Euclidean distance, the similarity between the user α and the user β is [(1−7) ² + (10−5) ² ] ^1/2 = 7.8. A feature space with proficiency levels of three or more dimensions can be considered similarly.

  The application determination unit 308 determines a user group based on the similarity or distance calculated in this way. Since the users are more similar as the distance between users in the N-dimensional feature space is shorter, the group of users is determined from the viewpoint of whether or not the distance is short. The maximum number that can be grouped is equal to the total number of users. From the viewpoint of simplifying the calculation by reducing the dimension of the N-dimensional feature space, it is preferable to convert components having high correlation into principal components. For example, the component “number of applications held in the communication terminal” and the component “storage capacity used in the communication terminal” are considered to be highly correlated with each other, so instead of these two components, It is preferable to consider one main component that combines these.

  The application determination unit 308 determines to which group of users the application is often used based on the grouping classification result, and determines the correspondence between the application and the group. For example, assume that the distribution of users having a certain application A is as shown in FIG. 4A. In this case, many users with low proficiency levels are distributed, and it can be seen that this application A is suitable for low proficiency level users, that is, beginner groups. The application determination unit 308 determines that the application A corresponds to the beginner group. A more detailed discrimination method will be described later. Information indicating that application A corresponds to the beginner's group (correspondence information) is typically stored in a storage unit in the distribution server or a storage unit other than the distribution server. It may be included in the metadata. As described above, the application determination unit 308 determines which of a plurality of groups an application corresponds to, and stores the correspondence relationship in a storage unit or metadata.

  The recommendation candidate determination unit (recommendation candidate determination unit) 310 in FIG. 3 creates a master list in which application candidates recommended for a specific user are listed based on the usage history information received from each user. As an example, the master list is a list that lists applications that a specific user α has not downloaded to his or her communication terminal. This list preferentially includes applications downloaded by the user β similar to the specific user α to his or her (user β) communication terminal.

  For example, the user α has downloaded the applications 1, 2, 3, 4 to his communication terminal, the user β has downloaded the applications 1, 2, P, Q to his communication terminal, and the user γ has himself Suppose that applications V and W have been downloaded to the communication terminal. First, the similarity between users is determined. The user β owns the same applications 1 and 2 as the user α, but the user γ does not own any of the same applications as the user α. Therefore, it can be seen that the users α and β have similar preference tendencies (preferences), and both are similar, but the users α and γ are not similar. In this way, among the applications owned by the user β, the applications P and Q not owned by the user α are determined as candidates recommended for the user α.

  The filtering unit 312 selects (filters) an application that is actually recommended to a specific user from candidates listed in the master list. The filtering 312 determines an application recommended for the specific user α by discriminating an application corresponding to a group (for example, a beginner group) to which the specific user α belongs, based on the above correspondence.

<4. Outline of operation>
FIG. 5 is a diagram for explaining the outline of the operation according to the embodiment of the present invention. The distribution server (not shown in FIG. 5) obtains usage history information from various users, calculates the proficiency level of each user from the usage history information, and classifies the user into three groups of beginner, intermediate, and advanced. The distribution server selects a plurality of users having a specific application A, and determines the correspondence between the specific application A and a specific group by examining how the selected users are distributed with respect to the proficiency level. To do. The distribution server that has received the recomment request signal from the specific user P determines the application recommended for the specific user P by determining the application corresponding to the group to which the specific user P belongs by the correspondence relationship. A recommendation signal recommending an application is transmitted to a specific user P. In the illustrated example, the user P belongs to the beginner's group, and an application associated with the beginner's group is recommended to the user P. The applications corresponding to the list presented to the user P are generally easy to operate, and many applications can be used immediately even by beginners. On the other hand, the user Q belongs to the intermediate group, and an application associated with the intermediate group is recommended to the user Q. The applications included in the list presented to the user Q are generally somewhat complicated to operate and many can only be used effectively if they are more experienced than beginners. Thus, according to the present embodiment, the distribution server can recommend an application to the user in consideration of the user's proficiency level.

<5. Example of operation>
FIG. 6 is a sequence diagram illustrating an operation example in the embodiment. This operation example is executed by the communication terminal 10 and the distribution server 30.

  In step S61, the communication terminal 10 collects usage history information. The usage history information includes one or more of the information on the operation amount of the communication terminal, the information on the resource usage, and the information on the application to be used, or other information. The usage history information is notified to the proficiency level calculation unit 306 of the distribution server 30. The usage history information may be reported to the distribution server 30 at a predetermined frequency, or may be reported in response to a request from a user, a distribution server, or the like.

  In step S62, the user's proficiency level is calculated. The proficiency level is determined based on one or more of various amounts included in the usage history information.

  In step S63, the filtering unit 312 of the distribution server 30 examines the relationship between the proficiency level and the number of users for each application, and associates a specific application with a specific group. First, a user having a specific application A is selected. More precisely, users who have a track record of using a specific application A are selected. A distribution function (y = f (x)) representing how the selected users are distributed with respect to the proficiency level x is determined. Thereby, it is possible to determine which group a particular application A corresponds to among groups of various proficiency levels.

  For example, it is assumed that the distribution function for the user having the application A has a shape as shown in FIG. 4A. In this case, since the number of users is large in the low proficiency level area (beginner group), application A is associated with the beginner group.

  It is assumed that the distribution function for the user having the application A has a shape as shown by the solid line in FIG. In this case, since the number of users is large in the area of high proficiency (advanced group), application A is associated with the advanced group.

  Assume that the distribution function for the user having the application A has a shape as shown by a broken line in FIG. In this case, since the number of users is large in the medium proficiency region (intermediate group), the application A is associated with the intermediate group.

  In the method described with reference to FIGS. 4A and 7, the application A is associated with the group having the largest number of users among the elementary, intermediate, and advanced groups. However, the group with the largest number of users is not always appropriate.

  For example, it is assumed that the distribution function for the user having the application A has a shape as shown in FIG. In this case, the number of users is the highest in the high proficiency area (advanced group). Therefore, according to the method described with reference to FIGS. 4A and 7, the application A corresponds to the advanced group. However, in the case of the example shown in FIG. 8, a user having a proficiency level of intermediate or higher indicates that the application A is used, and when recommending the application A, it is better to recommend it to an intermediate person than an advanced person. Good. This is because those who are less proficient usually require recommendation. From such a viewpoint, it is preferable that the application A is associated with “a group at a point where the number of users exceeds a predetermined threshold” instead of “a group with the largest number of users”. The predetermined threshold may be a value such as 30% of the total number of users.

  In any case, the filtering unit 312 uses a predetermined rule according to the shape of the distribution function for the application A so that the application A can be associated with some group. The predetermined rule is, for example, the discrimination method described with reference to FIGS. 4A, 7, and 8. However, the predetermined rule is not limited thereto, and any appropriate rule that can identify the group from the distribution function may be used. .

For example, the application determination unit 308 of the distribution server 30 calculates a usage rate for each group for a specific application, and specifies a specific application according to a comparison result between the distribution of the usage rates in a plurality of groups and a predetermined threshold. It may be associated with a group. For example, the ratio SLA of the number of users having a certain application A among the total number of users SL belonging to the elementary group is determined as the usage rate rL in the elementary group of the application A:
rL = SLA / SL.

Similarly, the ratio of the number SMA of users having a certain application A among the total number of users SM belonging to the intermediate group is determined as the usage rate rM in the intermediate group of application A:
rM = SMA / SM.

Similarly, the ratio of the number SUA of users having a certain application A out of the total number of users SU belonging to the advanced group is determined as the usage rate rU in the advanced group of application A:
rU = SUA / SU.
In this case, SL + SM + SU is equal to the total number of users N.

  A group associated with the application A is determined based on how the usage rate of each group for the application A thus calculated is distributed (based on the usage rate distribution). For example, when the variance calculated from the usage rates rL, rM, and rU is small, for example, when the usage rates rL, rM, and rU have similar values, the application A is widely used by users of various proficiency levels. Therefore, it is preferable to correspond to the beginner group. If the variance is large, for example if the utilization rU is very large and rL and rM are very small, the application is widely used in the advanced group but not so much in the beginner and intermediate groups. In this case, the application A is associated with the advanced group.

To summarize the above method, this method calculates the usage rate for each group for a specific application A, calculates the variance s for the usage rate for each group, and determines the group according to the size of the variance s. To do. For example, it is assumed that rL = rM = rU = 0.3 and the threshold value t is 0.04. In this case, the variance s is 0 and is smaller than the threshold value t. Therefore, application A is associated with the beginner group. In contrast, for example, it is assumed that rL = rM = 0.1 and rU = 0.7 (average is 0.3), and the threshold t is 0.04. In this case, the variance s is [(0.1−0.3) ² + (0.1−0.3) ² + (0.7−0.3) ² ] /3=0.08, which is a threshold value. Greater than. In this case, the application A is associated with an advanced group having the highest usage rate.

  The processing from step S61 to step S63 in FIG. 6 proceeds regardless of the user's intention, and each of the various applications existing in the market is associated with an appropriate group, and the correspondence information is stored in the server or metadata. Saved in.

  When the user of the communication terminal 10 operates the communication terminal 10, some screen is displayed on the display unit 210 of the communication terminal 10. Application recommendation is performed through this screen. In order to execute this, when the user α displays some web page on the display unit 210 in step S <b> 64, the communication terminal 10 transmits a recommendation request signal to the distribution server 30. The recommendation request signal is a trigger for recommending the application according to the present embodiment.

  In step S65, the recommendation candidate determination unit 310 of the distribution server 30 determines application candidates (recommendation candidates) recommended to the user α. This candidate is included in the above master list, and the master list is owned by the user β similar to the user α among the applications that the specific user α has not downloaded to his / her communication terminal. Are enumerated. Information on recommendation candidates included in the master list is notified from the recommendation candidate determination unit 310 to the filtering unit 312.

  In step S66, the filtering unit 312 determines an application to be actually recommended by determining an application corresponding to the group to which the user α belongs among the recommendation candidates. The recommendation candidates generally include an application corresponding to an elementary group, an application corresponding to an intermediate group, and an application corresponding to an advanced group. Although it is assumed that it is divided into three groups, the number of groups is not limited. The filtering unit 312 selects an application corresponding to a group to which the user α belongs (for example, a beginner group) from recommendation candidates, and determines the selected application as an application actually recommended to the user α. A recommendation signal indicating an application determined to be recommended is transmitted from the distribution server 30 to the communication terminal 10.

  In step S67, the communication terminal 10 displays, on the display unit 210, a list of applications that are recommendation candidates and correspond to the beginner group according to the recommendation signal. Thereby, an application suitable for the proficiency level of the user α of the communication terminal 10 can be appropriately recommended to the user α.

<6. Modification>
In the above method, users are divided into a plurality of groups according to their proficiency level. However, such grouping is not essential. That is, grouping may not be performed when recommending an application to a user in consideration of the user's proficiency level.

  FIG. 9 is a sequence diagram illustrating an operation example in the modification. This operation example is executed by the communication terminal 10 and the distribution server 30. Although generally the same as the sequence of FIG. 6, the processes in at least steps S93 and S96 are different.

  In step S61, the communication terminal 10 collects usage history information.

  In step S62, the user's proficiency level is calculated. The proficiency level is determined based on one or more of various amounts included in the usage history information.

  In step S63, the filtering unit 312 of the distribution server 30 checks the relationship between the proficiency level and the number of users for each application. First, a user having a specific application A is selected. More precisely, users who have a track record of using a specific application A are selected. A distribution function (y = f (x)) representing how the selected users are distributed with respect to the proficiency level x is determined.

  The processing up to Steps S61, S62, and S93 proceeds regardless of the user's intention, a distribution function is determined for each of various applications existing in the market, and distribution function information is stored in the distribution server.

  In step S <b> 64, when the user displays some web page on the display unit 210, the communication terminal 10 transmits a recommendation request signal to the distribution server 30.

  In step S65, the recommendation candidate determining unit 310 of the distribution server 30 determines application candidates to be recommended to the user. This candidate is included in the master list. Information on recommendation candidates included in the master list is notified from the recommendation candidate determination unit 310 to the filtering unit 312.

  In step S <b> 96, the filtering unit 312 determines an application that is actually recommended by determining an application whose number of users is equal to or greater than a threshold for the user's proficiency among the recommendation candidates.

FIG. 10 shows the relationship y = f _A (x), f _B (x) between the proficiency level x and the number y of users having the proficiency level for specific applications A and B included in the recommendation candidates. Show.

When receiving the recommendation request signal from the user P, the filtering unit 312 determines whether or not the number of users in the proficiency level of the user P exceeds a predetermined threshold based on the distribution function y = f _A (x) related to the application A. Is determined. In the illustrated example, the number of users exceeds the threshold. Therefore, the application A is determined as an application that is actually recommended to the user P. Further, the filtering unit 312 determines whether or not the number of users in the proficiency level of the user P exceeds a predetermined threshold based on the distribution function y = f _B (x) related to the specific application B. In the illustrated example, the number of users does not exceed the threshold. Therefore, it is determined that the application B is not actually recommended to the user P.

If a recommendation request signal is received from a user Q who has a different level of proficiency than the user P, the filtering unit 312 determines that the user at the proficiency level of the user Q is based on the distribution function y = f _A (x) related to the application A. It is determined whether or not the number exceeds a predetermined threshold. In the illustrated example, the number of users does not exceed the threshold. Therefore, it is determined that the application A is not actually recommended to the user Q. Further, the filtering unit 312 determines whether the number of users in the proficiency level of the user Q exceeds a predetermined threshold based on the distribution function y = f _B (x) related to the specific application B. In the illustrated example, the number of users exceeds the threshold. Therefore, the application B is determined as an application that is actually recommended to the user Q.

  In this way, the filtering unit 312 determines whether or not the number of users in the proficiency level of the target users P and Q exceeds a predetermined threshold by using the distribution function of each application in the recommendation candidates. Thus, the application actually recommended to the users P and Q is determined. A recommendation signal indicating an application determined to be recommended is transmitted from the distribution server 30 to the communication terminal 10.

  In step S67, the communication terminal 10 displays a list of applications determined to be recommended on the display unit 210 according to the recommendation signal. By doing in this way, the application suitable for the user's proficiency level of the communication terminal 10 can be appropriately recommended to the user.

  Although the present invention has been described with reference to particular embodiments, they are merely exemplary and those skilled in the art will appreciate various variations, modifications, alternatives, substitutions, and the like. For example, the present invention may be applied to any appropriate mobile communication system that recommends applications to users. Although specific numerical examples have been described in order to facilitate understanding of the invention, these numerical values are merely examples and any appropriate values may be used unless otherwise specified. Although specific mathematical formulas have been described to facilitate understanding of the invention, these mathematical formulas are merely examples, unless otherwise specified, and any appropriate mathematical formula may be used. The classification of the examples or items is not essential to the present invention, and the items described in two or more items may be used in combination as necessary, and the items described in one item may be combined with other items. It may be applied (as long as it is not inconsistent) to the matters described. For convenience of explanation, an apparatus according to an embodiment of the present invention has been described using a functional block diagram, but such an apparatus may be realized by hardware, software, or a combination thereof. The software is available on random access memory (RAM), flash memory, read only memory (ROM), EPROM, EEPROM, registers, hard disk (HDD), removable disk, CD-ROM, database, server and any other suitable storage medium May be. The present invention is not limited to the above embodiments, and various modifications, modifications, alternatives, substitutions, and the like are included in the present invention without departing from the spirit of the present invention.

  Hereinafter, the means taught by the embodiments of the present invention will be exemplified.

(A1)
A receiving unit that receives usage history information indicating the usage history of the application from a plurality of users;
A proficiency level calculating unit for calculating proficiency levels of each of a plurality of users having a specific application from the usage history information;
Classifying the plurality of users having the specific application into two or more groups according to the proficiency level of each of the users, and determining an association that associates the specific application with any group; and
A filtering unit that determines an application recommended for the specific user by determining an application corresponding to a group to which the specific user belongs by the correspondence relationship;
A distribution server for recommending an application to a user, comprising: a transmission unit that transmits a recommendation signal for recommending an application determined to be recommended by the filtering unit to the specific user.

(A2)
The distribution server according to A1, wherein the application determination unit classifies the plurality of users into two or more groups according to the proficiency level of each user by using a cluster analysis method.

(A3)
The distribution server according to A1 or A2, wherein the application determination unit classifies the plurality of users into three groups of an elementary level, an intermediate level, and an advanced level according to the proficiency level of each user.

(A4)
The application determination unit calculates a usage rate for each group for a specific application, and associates the specific application with a specific group according to a comparison result between the distribution of the usage rates in a plurality of groups and a predetermined threshold value. The delivery server according to any one of A1 to A3.

(A5)
By determining a similar user whose application usage tendency is similar to that of a specific user, and determining an application owned by the similar user among the applications not owned by the specific user, the specific user is identified. It further has a recommendation candidate determination unit for determining candidate applications to be recommended,
The filtering unit determines an application recommended for the specific user by determining an application corresponding to the group to which the specific user belongs among the candidates for the application recommended for the specific user based on the correspondence relationship. The delivery server according to any one of A1 to A4 to be determined.

(A6)
The usage history information includes the time when the user used the communication terminal, the number of times the user operated through the user interface, the number of applications stored in the communication terminal, the CPU usage of the communication terminal, the use of the storage unit of the communication terminal The distribution server according to any one of A1 to A5, including information indicating one or more of the amount, the number of applications used by the user, and the number of times the application is activated.

(A7)
A method of recommending an application to a user by a distribution server,
Receive usage history information indicating the usage history of the application from multiple users,
A proficiency level of each of a plurality of users having a specific application is calculated from the usage history information,
Classifying the plurality of users having the specific application into two or more groups according to the proficiency level of each user, and determining a correspondence relationship that associates the specific application with any group;
By determining an application corresponding to a group to which a specific user belongs based on the correspondence relationship, an application recommended for the specific user is determined,
A method of recommending an application to a user, comprising: transmitting a recommendation signal recommending an application decided to be recommended to the specific user.

(B1)
A receiving unit that receives usage history information indicating the usage history of the application from a plurality of users;
A proficiency level in which a proficiency level of each of a plurality of users having a specific application is calculated from the use history information, and a distribution function representing a relationship between the proficiency level and the number of users having the proficiency level is determined for each application. A calculation unit;
By determining whether or not the number of users corresponding to the proficiency level of a specific user exceeds a predetermined value based on the distribution function for the specific application, the specific application is set as an application recommended to the specific user. A filtering unit for determining whether or not
A distribution server for recommending an application to a user, comprising: a transmission unit that transmits a recommendation signal for recommending an application determined to be recommended by the filtering unit to the specific user.

(B2)
By determining a similar user whose application usage tendency is similar to that of a specific user, and determining an application owned by the similar user among the applications not owned by the specific user, the specific user is identified. It further has a recommendation candidate determination unit for determining candidate applications to be recommended,
The filtering unit specifies a specific application among candidate applications recommended for the specific user, and specifies whether the number of users corresponding to the proficiency level of the specific user exceeds a predetermined value. The distribution server according to B <b> 1, which determines whether or not the specific application is an application recommended to the specific user by determining the distribution function for the specific application.

(B3)
The usage history information includes the time when the user used the communication terminal, the number of times the user operated through the user interface, the number of applications stored in the communication terminal, the CPU usage of the communication terminal, the use of the storage unit of the communication terminal The distribution server according to B1 or B2, including information indicating one or more of the amount, the number of applications used by the user, and the number of times the application is activated.

(B4)
A method of recommending an application to a user by a distribution server,
Receive usage history information indicating the usage history of the application from multiple users,
A proficiency level of each of a plurality of users having a specific application is calculated from the usage history information, and a distribution function representing a relationship between the proficiency level and the number of users having the proficiency level is determined for each application.
By determining whether or not the number of users corresponding to the proficiency level of a specific user exceeds a predetermined value based on the distribution function for the specific application, the specific application is set as an application recommended to the specific user. Whether or not
A method of recommending an application to a user, comprising: transmitting a recommendation signal recommending an application decided to be recommended to the specific user.

DESCRIPTION OF SYMBOLS 1 Communication system 10 Communication terminal 20 Mobile communication network 30 Distribution server 40 Application management server 202 Transmission / reception part 204 Terminal operation management part 206 Resource usage management part 208 Application management part 210 Display part 302 Transmission / reception part 304 Usage history information storage part 306 Calculation unit 308 Application determination unit 310 Recommendation candidate determination unit 312 Filtering unit

Claims (8)

A receiving unit that receives usage history information indicating the usage history of the application from a plurality of users;
And proficiency calculation unit that calculates a learning level of the plurality of users each from the previous SL usage history information,
Classified into one of two or more groups according to the plurality of users each on their proficiency, by calculating the number of users for each learning level for each application, two of the plurality of applications each an application determination unit that determines a correspondence relationship indicating which corresponds to any of the above groups,
The applications that are corresponding to the group which a particular user belongs among the two or more groups, by determining by said correspondence relationship, a filtering unit that determines an application to be recommended to the particular user,
A distribution server for recommending an application to a user, comprising: a transmission unit that transmits a recommendation signal for recommending an application determined to be recommended by the filtering unit to the specific user.   The distribution server according to claim 1, wherein the application determination unit classifies the plurality of users into two or more groups according to a proficiency level of each user by using a cluster analysis method. The distribution server according to claim 1, wherein the application determination unit classifies the plurality of users into three groups of a beginner level, an intermediate level, and an advanced level according to the proficiency level of each user. The application determination unit calculates a usage rate of each group for a plurality of applications each, in accordance with the comparison result between the dispersion with a predetermined threshold value of the usage rate of the plurality of groups, that determine the relationship, claim 4. The distribution server according to any one of 1 to 3. The distribution server determines a similar user whose application usage tendency is similar to a specific user, and determines an application owned by the similar user among applications not owned by the specific user. A recommendation candidate determination unit that determines application candidates to be recommended to the specific user;
The filtering unit, in the candidates of the application to be recommended to the particular user, determines the corresponding said particular users that correspond to the groups belonging in relation application as the application to be recommended before Symbol particular user The delivery server according to any one of claims 1 to 4.   The usage history information includes the time when the user used the communication terminal, the number of times the user operated through the user interface, the number of applications stored in the communication terminal, the CPU usage of the communication terminal, the use of the storage unit of the communication terminal The distribution server according to claim 1, comprising information indicating one or more of the amount, the number of applications used by the user, and the number of times the application has been activated. A recommendation method in which delivery server that recommend applications to users run,
Receiving the usage history information indicating an application utilization history from a plurality of users,
Calculating a learning level of the plurality of users each from the previous SL usage history information,
Classified into one of two or more groups according to the plurality of users each on their proficiency, by calculating the number of users for each learning level for each application, two of the plurality of applications each determining a correspondence indicating whether corresponding to any of the above groups,
The applications that are corresponding to the group which a particular user belongs among the two or more groups, by determining by said correspondence relationship, a filtering step of determining an application to be recommended to the particular user,
The recommendation signal to recommend the application determined to be recommended by the filtering step, and a step of transmitting to the particular user, the recommendation how.   The distribution server determines a similar user whose application usage tendency is similar to that of a specific user, and determines an application owned by the similar user among applications not owned by the specific user, A recommendation candidate determination step for determining a candidate for an application recommended for the specific user;
  In the filtering step, among the candidate applications recommended for the specific user, an application corresponding to the group to which the specific user belongs in the correspondence relationship is determined as an application recommended for the specific user. The recommendation method according to claim 7.

Images